Project Summary The Filoviridae family consists of multiple phylogenetically diverse species including five species of ebolavirus and a single marburgvirus species with Marburg and Ravn viruses. While Ebola virus (EBOV; formerly known as Zaire) has caused the majority of filovirus hemorrhagic fever epidemics including the 2014 outbreak in West Africa, other members of Filoviridae have also caused human epidemics including 12, 8, and 2 outbreaks of Marburg (MARV), Sudan (SUDV), and Bundibugyo (BDBV) viruses, respectively. It is impossible to predict the species or location of future outbreaks, thus a broadly protective therapeutic would be highly desirable. However, current immunotherapeutic candidates including ZMapp are species-specific. The objective of this project is to develop broadly neutralizing pan-filovirus antibodies with protection against the major filovirus species. Integrated Biotherapeutics Inc. (IBT) has recently developed a set of pan-ebolavirus monoclonal antibodies that target novel conserved epitopes within filovirus glycoproteins (GP). Two of these mAbs bind with low affinity to MARV, which is phylogenetically distant and has only 28% sequence identity within GP with ebolavirus species. Visterra Inc, has developed a framework to compute the inter-residue atomic interactions between interacting amino acid pairs of an antigen-antibody interface. This technology enables in silico optimization of CDRs to enhance amino acid interaction fitness at the paratope-epitope interface to drive increased affinity or widening of the breadth of reactivity. Using this approach a Dengue virus mAb was engineered to achieve 13000 fold increased binding to DENV-IV and a pan-influenza mAb was created that is now in clinical development. In this proposal, IBT and Visterra are partnering to use this novel approach to create effective pan-filovirus mAbs by expanding the breadth of reactivity of two antibodies that bind with high affinity to novel pan-ebolavirus epitopes within the core GP1 and the fusion loop of all ebolavirus species but show low affinity towards MARV. In Specific Aim 1, these two mAbs will be subjected to an iterative structure-based engineering to achieve strong binding to Marburg glycoprotein without loss of ebolavirus reactivity. Up to 10 derivatives with broad cross reactivity to all filoviruses will be selected for further characterization and efficacy testing. In Aim 2, the selected redesigned mAbs will be thoroughly characterized for cross-species binding, affinity, as well as neutralization. The most potent antibodies (or cocktails) will be evaluated in mouse models of EBOV, SUDV, and MARV. If the Phase I is successful, a Phase II project is envisioned in which the efficacy of the candidates will be tested in nonhuman primate models of EBOV, SUDV, MARV, and BDBV with the ultimate goal of advancing the product towards clinical development.